EuroEval Spanish NLU - ScaLA ES — leaderboard
Metric: Linguistic acceptability Score (%). Source: euroeval.com. 413 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 73.71 |
| 2 | Claude Sonnet 4.6 | 69.02 |
| 3 | Gemini 3 Flash (Preview) (Non-reasoning) | 68.43 |
| 4 | GPT-5.4 Mini (High) | 62.38 |
| 5 | GPT-5.2 | 61.92 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 61.41 |
| 7 | GPT-5.4 Mini (Non-reasoning) | 60.47 |
| 8 | Llama 3.1 405B | 56.52 |
| 9 | GPT-5.4 Mini (Medium) | 56.45 |
| 10 | GPT-5.4 Nano (High) | 55.08 |
| 11 | Qwen 3.5 9B | 52.26 |
| 12 | Grok 4.1 Fast (Reasoning) | 52.16 |
| 13 | GPT-5.4 Nano (Medium) | 49.61 |
| 14 | Qwen 3 4B 2507 (Thinking) | 49.21 |
| 15 | Ministral-3-3B-Instruct-2512 | 47.08 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-spanish-nlu-scala-es · How the rankings work · Data refreshed daily, snapshot 2026-07-22.